Machine learning Based ASD classification using structural MRI
Budget: $20 – $125 USD
I need someone skilled who has worked in machine learning applications before on MRI data To build a brain disorder diagnosis using preprocessed structural magnetic resonance images from the ABIDE suite, manually extracted features and automatically extracted features are combined in the diagnostic process and some demographic information is taken into account. An example of the work :https://www.spiedigitallibrary.org/conference-proceedings-of-spie/11314/113141W/Combining-deep-and-hand-crafted-MRI-features-for-identifying-sex/10.1117/12.2551341.short?SSO=1
As I mentioned, they are two models, one of which applies SVM or any other conventional ML classifier to extract the features and the other applies CNN, then the results of the two classifiers are combined to make the final decision
Also I need the gradient-CAM attention unit mode with the CNN model to clarify the important areas on which the classifier's decision depends
need an precise model building, in Python3 only (spyder, jupyter ,etc)well documented
This data set contains more than 1000 images, while I only need 200 from the age of 5 to 11
,There are alot of demographic and clinical data with each subject in excel file , however only I need sex to be trained on the model (as the discriminative features between two classes of patients and healthy control , also different between male and female patients
for the evaluation : I need 10 cross-validation and Accuracy,sensitivty, specificity measures
I am going to share with you a paper that published a code that uses the traditional machine learning technique using the morphological brain network feature and the same dataset that we will be using
As I mentioned, they are two models, one of which applies SVM or any other conventional ML classifier to extract the features and the other applies CNN, then the results of the two classifiers are combined to make the final decision
Also I need the gradient-CAM attention unit mode with the CNN model to clarify the important areas on which the classifier's decision depends
need an precise model building, in Python3 only (spyder, jupyter ,etc)well documented
This data set contains more than 1000 images, while I only need 200 from the age of 5 to 11
,There are alot of demographic and clinical data with each subject in excel file , however only I need sex to be trained on the model (as the discriminative features between two classes of patients and healthy control , also different between male and female patients
for the evaluation : I need 10 cross-validation and Accuracy,sensitivty, specificity measures
I am going to share with you a paper that published a code that uses the traditional machine learning technique using the morphological brain network feature and the same dataset that we will be using